5.1. Collections¶
In addition to the numeric, Boolean, and character types, C++ also offers built-in collection types. A collection data type is a grouping of some number of other data items (possibly only zero or one) that have some shared significance or need to be operated upon together.
Arrays, vectors, strings, sets, and hash tables are among these useful C++ collection types.
5.2. Arrays¶
What is an Array? An array is a data structure consisting of an ordered collection of data elements, of identical type in which each element can be identified by an array index. More technically, an array data structure is an ordered arrangement of values located at equally spaced addresses in contiguous computer memory.
NOTE: A C++ array is always stored in contiguous memory. C++ arrays can be allocated in two different ways:
statically allocated in which the array size is fixed at compile-time and cannot change
dynamically allocated in which pointers are used in the allocation process so the size can change at run-time
In modern C++, the statically allocated array is typically used in situations when speed is essential or where hardware constraints exist, and a data structure called a vector is typically used when more flexibility is required.
As a Python programmer, you might see the array as the ancestor of the Python list, and you might remember that Python lists are actually implemented via an underlying array consisting of references.
C++ arrays are similar to Python lists except that because C++ stores variables directly, each element of a C++ array must be of identical data type. Like Python lists, the indices for arrays start counting with 0.
The use of arrays permits us to utilize an ordered set of memory locations that we can then manipulate as a single entity, and that at the same time gives us direct access to each individual component.
Why use an Array?
C++ is a language often used for real-time or low-level processing where speed is essential and/or there is only a fixed amount of space available.
The fact that array elements are stored in memory in contiguous memory locations making look-up via index very, very fast. In computing, a word is the unit of data used by a particular processor design, such as 32 or 64 bits. For example, an array of 100 integer variables, with indices 0 through 99, might be stored as 100 words at memory addresses 20000, 20004, 20008, … 20396. The element with index i would be located at the address 20000 + 4 × i.
Statically allocated C++ arrays must be declared with both a type and a size at compile-time.
double darray[4];
int iarray[10];
char arr2[3000];
It is also possible to initialize statically allocated arrays at compile time, in which case the number of items determines the array’s size.
int arr[] = {1, 2, 3, 4}; // This is size 4.
char arr2[] = {'a', 'b', 'c'}; // This is size 3.
string arr3[] = {"this", "is", "an", "array"}; // This is size 4.
Note that an array with a single element is not the same type as the atomic type, so the following are not the same.
double darray[] = {1.2}; // must use index to access value
double ddouble = 1.2; // cannot use index to access value
Be Cautious with Arrays
The speed and low-level control that arrays offer us as programmers is powerful… and dangerous. As a Python programmer, using a C++ array will help you better understand the trade-offs of the protections Python offers.
Here are examples of iteration.
The protections Python offers, however, takes time and C++ is designed for speed. Python would never let you iterate beyond the end of a list. C++ will not only let you iterate beyond either end of an array, but it will let you change the values beyond either end of the array with sometimes catastrophic results.
The phrase, “be careful what you wish for” is a great one to remember when programming in C++. Because C++ will generally try to do everything you ask for.
The speed of C++ comes at the cost of minimal to no error checking. Sometimes this can have perplexing results such as in the next example.
You should use an array when you have a need for speed or you need to work with hardware constraints. Otherwise, you may want to consider using another collection data type, the vector.
- Nothing. Everything is fine.
- Actually, there is a problem. Look carefully.
- All data was automatically reinitialized.
- No. C++ just does what you tell it to do.
- I have no idea. Please give me a hint.
- Try again. One of these is indeed correct. Look at the memory addresses.
- The first loop went out of bounds and wrote over the values in otherdata.
- Right!
- none of the above
- One of the above is indeed correct.
Q-7: In the above example, what happened to otherdata[ ] in C++?
- int myarray(5);
- Check the characters at the end of the array! Right now that is a function!
- myarray[5];
- You are forgetting something important!
- int myarray[5];
- Good work!
- None of the above.
- Check the characters at the end of the array!
Q-8: What is the correct way to declare an array in C++?